US20150193816A1 - Methods and apparatus to correct misattributions of media impressions - Google Patents

Abstract

An example method involves determining an impressions adjustment factor for a first demographic group based on first impressions reported by a client device to a first internet domain and second impressions reported by the client device to a second internet domain. The first and second impressions correspond to same media accessed on the client device. The example also involves determining a misattribution-corrected impressions count for the first demographic group based on the impressions adjustment factor and based on a second impressions count determined at the second internet domain for the first demographic group. The second impressions count has an error based on some of the second impressions being misattributed at the second internet domain to the first demographic group when the some of the second impressions correspond to a second demographic group.

Description

RELATED APPLICATION
• This application claims priority to U.S. Provisional Patent Application Ser. No. 61/923,959 filed on Jan. 6, 2014, which is hereby incorporated herein by reference in its entirety.
FIELD OF THE DISCLOSURE
• The present disclosure relates generally to monitoring media and, more particularly, to methods and apparatus to correct misattributions of media impressions.
BACKGROUND
• Traditionally, audience measurement entities determine audience engagement levels for media based on registered panel members. That is, an audience measurement entity enrolls people who consent to being monitored into a panel. The audience measurement entity then monitors those panel members to determine media (e.g., television programs or radio programs, movies, DVDs, advertisements, streaming media, websites, etc.) exposed to those panel members. In this manner, the audience measurement entity can determine exposure metrics for different media based on the collected media measurement data.
• Techniques for monitoring user access to Internet resources such as web pages, advertisements and/or other Internet-accessible media have evolved significantly over the years. Some known systems perform such monitoring primarily through server logs. In particular, entities serving media on the Internet can use known techniques to log the number of requests received for their media (e.g., content and/or advertisements) at their server.
BRIEF DESCRIPTION OF THE DRAWINGS
• FIG. 1 illustrates an example client device that reports audience impressions for media to impression collection entities to facilitate identifying total impressions and sizes of unique audiences exposed to different media.
• FIG. 2 is an example communication flow diagram of an example manner in which an audience measurement entity (AME) and a database proprietor (DP) can collect impressions and demographic information based on a client device reporting impressions to the AME and the DP.
• FIG. 3 illustrates example impressions collected by the AME and example impressions collected by the DP with a misattribution error.
• FIG. 4 illustrates example audience adjustment (AA) factors for unique audience sizes of different demographic groups determined based on the example impressions ofFIG. 3.
• FIG. 5 illustrates example impression adjustment (IA) factors for total impressions of different demographic groups determined based on the example impressions ofFIG. 3.
• FIG. 6 illustrates example misattribution-corrected unique audience values and example misattribution-corrected impression counts determined based on the example AA factors ofFIG. 4 and the example IA factors ofFIG. 5 for different demographic groups.
• FIG. 7 illustrates example misattribution-corrected unique audience values and example misattribution-corrected impression counts determined based on the example IA factors ofFIG. 5 and example impression frequencies for different demographic groups.
• FIG. 8 is a flow diagram representative of example machine readable instructions that may be executed to implement the misattribution corrector ofFIG. 2 to determine misattribution-corrected unique audience sizes and misattribution-corrected impression counts.
• FIG. 9 illustrates an example processor system structured to execute the example instructions ofFIG. 8 to implement the example AME ofFIGS. 1 and/or2.
DETAILED DESCRIPTION
• Techniques for monitoring user access to Internet-accessible media such as web pages, advertisements, content and/or other media have evolved significantly over the years. At one point in the past, such monitoring was done primarily through server logs. In particular, entities serving media on the Internet would log the number of requests received for their media at their server. Basing Internet usage research on server logs is problematic for several reasons. For example, server logs can be tampered with either directly or via zombie programs which repeatedly request media from the server to increase the server log counts. Secondly, media is sometimes retrieved once, cached locally and then repeatedly viewed from the local cache without involving the server in the repeat viewings. Server logs cannot track these repeat views of cached media. Thus, server logs are susceptible to both over-counting and under-counting errors.
• The inventions disclosed in Blumenau, U.S. Pat. No. 6,108,637, fundamentally changed the way Internet monitoring is performed and overcame the limitations of the server side log monitoring techniques described above. For example, Blumenau disclosed a technique wherein Internet media to be tracked is tagged with beacon instructions. In particular, monitoring instructions are associated with the hypertext markup language (HTML) of the media to be tracked. When a client requests the media, both the media and the beacon instructions are downloaded to the client. The beacon instructions are, thus, executed whenever the media is accessed, be it from a server or from a cache.
• The beacon instructions cause monitoring data reflecting information about the access to the media to be sent from the client that downloaded the media to a monitoring entity. Typically, the monitoring entity is an audience measurement entity (AME) that did not provide the media to the client and who is a trusted (e.g., neutral) third party for providing accurate usage statistics (e.g., The Nielsen Company, LLC). Advantageously, because the beaconing instructions are associated with the media and executed by the client browser whenever the media is accessed, the monitoring information is provided to the AME irrespective of whether the client is a panelist of the AME.
• Audience measurement entities and/or other businesses often desire to link demographics to the monitoring information. To address this issue, the AME establishes a panel of users who have agreed to provide their demographic information and to have their Internet browsing activities monitored. When an individual joins the panel, they provide detailed information concerning their identity and demographics (e.g., gender, age, ethnicity, income, home location, occupation, etc.) to the AME. The audience measurement entity sets a cookie on the panelist computer that enables the audience measurement entity to identify the panelist whenever the panelist accesses tagged media and, thus, sends monitoring information to the audience measurement entity.
• Since most of the clients providing monitoring information from the tagged media are not panelists and, thus, are unknown to the audience measurement entity, it is necessary to use statistical methods to impute demographic information based on the data collected for panelists to the larger population of users providing data for the tagged media. However, panel sizes of audience measurement entities remain small compared to the general population of users. Thus, a problem is presented as to how to increase panel sizes while ensuring the demographics data of the panel is accurate.
• There are many database proprietors operating on the Internet. These database proprietors provide services to large numbers of subscribers. In exchange for the provision of the service, the subscribers register with the proprietor. As part of this registration, the subscribers provide detailed demographic information. Examples of such database proprietors include social network providers, email providers, etc. such as Facebook, Myspace, Twitter, Yahoo!, Google, etc. These database proprietors set cookies or other device/user identifiers on the client devices of their subscribers to enable the database proprietor to recognize the user when they visit their website.
• The protocols of the Internet make cookies inaccessible outside of the domain (e.g., Internet domain, domain name, etc.) on which they were set. Thus, a cookie set, for example, in the amazon.com domain is accessible to servers in the amazon.com domain, but not to servers outside that domain. Therefore, although an audience measurement entity might find it advantageous to access the cookies set by the database proprietors, they are unable to do so.
• The inventions disclosed in Mainak et al., U.S. Pat. No. 8,370,489, which is incorporated by reference herein in its entirety, enable an audience measurement entity to leverage the existing databases of database proprietors to collect more extensive Internet usage and demographic data by extending the beaconing process to encompass partnered database proprietors and by using such partners as interim data collectors. The inventions disclosed in Mainak et al. accomplish this task by structuring the AME to respond to beacon requests from clients (who may not be a member of an audience member panel and, thus, may be unknown to the audience member entity) and redirect the client from the audience measurement entity to a database proprietor such as a social network site partnered with the audience member entity. The redirection initiates a communication session between the client accessing the tagged media and the database proprietor. The database proprietor (e.g., Facebook) can access any cookie it has set on the client to thereby identify the client based on the internal records of the database proprietor. In the event the client corresponds to a subscriber of the database proprietor, the database proprietor logs an impression in association with the demographics data associated with the client and subsequently forwards logged impressions to the audience measurement company. In the event the client does not correspond to a subscriber of the database proprietor, the database proprietor may redirect the client to the audience measurement entity and/or another database proprietor. The audience measurement entity may respond to the redirection from the first database proprietor by redirecting the client to a second, different database proprietor that is partnered with the audience measurement entity. That second database proprietor may then attempt to identify the client as explained above. This process of redirecting the client from database proprietor to database proprietor can be performed any number of times until the client is identified and the media exposure logged, or until all database proprietor partners have been contacted without a successful identification of the client. The redirections all occur automatically so the user of the client is not involved in the various communication sessions and may not even know they are occurring.
• Periodically or aperiodically the partnered database proprietors provide their logs and demographic information to the audience measurement entity which then compiles the collected data into statistical reports accurately identifying the demographics of persons accessing the tagged media. Because the identification of clients is done with reference to enormous databases of users far beyond the quantity of persons present in a conventional audience measurement panel, the data developed from this process is extremely accurate, reliable and detailed.
• Significantly, because the audience measurement entity remains the first leg of the data collection process (e.g., receives the request generated by the beacon instructions from the client), the audience measurement entity is able to obscure the source of the media access being logged as well as the identity of the media itself from the database proprietors (thereby protecting the privacy of the media sources), without compromising the ability of the database proprietors to log impressions for their subscribers. Further, when cookies are used as device/user identifiers, the Internet security cookie protocols are complied with because the only servers that access a given cookie are associated with the Internet domain (e.g., Facebook.com) that set that cookie.
• Examples disclosed in Mainak et al. (U.S. Pat. No. 8,370,489) can be used to determine any type of media impressions or exposures (e.g., content impressions, advertisement impressions, content exposure, and/or advertisement exposure) using demographic information, which is distributed across different databases (e.g., different website owners, service providers, etc.) on the Internet. Not only do such disclosed examples enable more accurate correlation of Internet advertisement exposure to demographics, but they also effectively extend panel sizes and compositions beyond persons participating in the panel of an audience measurement entity and/or a ratings entity to persons registered in other Internet databases such as the databases of social media sites such as Facebook, Twitter, Google, etc. Such extension effectively leverages the media tagging capabilities of the ratings entity and the use of databases of non-ratings entities such as social media and other websites to create an enormous, demographically accurate panel that results in accurate, reliable measurements of exposures to Internet media such as advertising and/or programming.
• In illustrated examples disclosed herein, media exposure is measured in terms of online Gross Rating Points. A Gross Rating Point (GRP) is a unit of measurement of audience size that has traditionally been used in the television ratings context. It is used to measure exposure to one or more media (e.g., programs, advertisements, etc.) without regard to multiple exposures of the same media to individuals. In terms of television (TV) advertisements, one GRP is equal to 1% of TV households. While GRPs have traditionally been used as a measure of television viewership, examples disclosed herein may be used in connection with generating online GRPs for online media to provide a standardized metric that can be used across the Internet to accurately reflect online advertisement exposure. Such standardized online GRP measurements can provide greater certainty to advertisers that their online advertisement money is well spent. It can also facilitate cross-medium comparisons such as viewership of TV advertisements and online advertisements, exposure to radio advertisements and online media, etc. Because examples disclosed herein may be used to correct impressions that associate exposure measurements with corresponding demographics of users, the information processed using examples disclosed herein may also be used by advertisers to more accurately identify markets reached by their advertisements and/or to target particular markets with future advertisements.
• Traditionally, audience measurement entities (also referred to herein as “ratings entities”) determine demographic reach for advertising and media programming based on registered panel members. That is, an audience measurement entity enrolls people that consent to being monitored into a panel. During enrollment, the audience measurement entity receives demographic information from the enrolling people so that subsequent correlations may be made between advertisement/media exposures to those panelists and different demographic markets. Unlike traditional techniques in which audience measurement entities rely solely on their own panel member data to collect demographics-based audience measurements, example methods, apparatus, and/or articles of manufacture disclosed herein enable an audience measurement entity to share demographic information with other entities that operate based on user registration models. As used herein, a user registration model is a model in which users subscribe to services of those entities by creating an account and providing demographic-related information about themselves. Sharing of demographic information associated with registered users of database proprietors enables an audience measurement entity to extend or supplement their panel data with substantially reliable demographics information from external sources (e.g., database proprietors), thus extending the coverage, accuracy, and/or completeness of the AMA's demographics-based audience measurements. Such access also enables the audience measurement entity to monitor persons who would not otherwise have joined an audience measurement panel. Any entity having a network-accessible database identifying demographics of a set of individuals may cooperate with the audience measurement entity. Such entities may be referred to as “database proprietors” and include entities such as Facebook, Google, Yahoo!, MSN, Twitter, Apple iTunes, Experian, etc.
• To increase the likelihood that measured viewership is accurately attributed to the correct demographics, examples disclosed herein use demographic information located in the audience measurement entity's records as well as demographic information located at one or more database proprietors that maintain records or profiles of users having accounts therewith. In this manner, examples disclosed herein may be used to supplement demographic information maintained by a ratings entity (e.g., an AME such as The Nielsen Company of Schaumburg, Ill., United States of America, that collects media exposure measurements and/or demographics) with demographic information from one or more different database proprietors.
• The use of demographic information from disparate data sources (e.g., high-quality demographic information from the panels of an audience measurement company and/or registered user data of web service providers) results in improved reporting effectiveness of metrics for both online and offline advertising campaigns. Example techniques disclosed herein use online registration data to identify demographics of users and use server impression counts, tagging (also referred to herein as beaconing), and/or other techniques to track quantities of impressions attributable to those users. Online web service providers such as social networking sites (e.g., Facebook) and multi-service providers (e.g., Yahoo!, Google, Experian, etc.) (collectively and individually referred to herein as database proprietors) maintain detailed demographic information (e.g., age, gender, geographic location, race, income level, education level, religion, etc.) collected via user registration processes. As used herein, an impression is defined to be an event in which a home or individual is exposed to corresponding media (e.g., content and/or an advertisement). Thus, an impression represents a home or an individual having been exposed to media (e.g., an advertisement, content, a group of advertisements, and/or a collection of content). In Internet advertising, a quantity of impressions or impression count is the total number of times media (e.g., content, an advertisement or advertisement campaign) has been accessed by a web population (e.g., the number of times the media is accessed). As used herein, a demographic impression is defined to be an impression that is associated with a characteristic (e.g., a demographic characteristic) of the person exposed to the media.
• Although such techniques for collecting media impressions are based on highly accurate demographic information, in some instances collected impressions may be misattributed to the wrong person and, thus, associated with incorrect demographic information. For example, in a household having multiple people that use the same client device (e.g., the same computer, tablet, smart internet appliance, etc.), collected impressions from that client device may be misattributed to a member of the household that is not the current user of the client device. That is, when an online user visits a website and is exposed to an advertisement on that site that has been tagged with beacon instructions, there is a redirect to a server of a database proprietor (e.g., Facebook, Yahoo, Google, etc.). The database proprietor then looks into a most recent cookie set by the database proprietor in the web browser of that client device. The database proprietor then attributes the impression to the user account corresponding to the cookie value. For example, the cookie value is one that was previously set in the client device by the database proprietor to correspond to a particular registered user account of the person that used the client device to most recently log into the website of that database proprietor. After collecting and attributing the impression to the user account associated with the retrieved cookie value, the database proprietor aggregates the total collected impressions and the size of the unique audience based on demographics associated with user accounts for all logged impressions. When this occurs over time and across many households, a number of collected impressions are misattributed to the wrong demographic information because some people use client devices after another person (e.g., another household member) has logged into a user account registered with the database proprietor without logging themselves (e.g., the current audience member) in. As such, a cookie corresponding to the previous person is still accessed from the client device while the subsequent user of the client device (e.g., a user that did not log into a corresponding user account registered with the database proprietor) accesses media on the client device which causes impressions to be misattributed to the previous person associated with the accessed cookie.
• As used herein, a unique audience measure is based on audience members distinguishable from one another. That is, a particular audience member exposed to particular media is measured as a single unique audience member regardless of how many times that audience member is exposed to that particular media. If that particular audience member is exposed multiple times to the same media, the multiple exposures for the particular audience member to the same media is counted as only a single unique audience member. In this manner, impression performance for particular media is not disproportionately represented when a small subset of one or more audience members is exposed to the same media an excessively large number of times while a larger number of audience members is exposed fewer times or not at all to that same media. By tracking exposures to unique audience members, a unique audience measure may be used to determine a reach measure to identify how many unique audience members are reached by media. In some examples, increasing unique audience and, thus, reach, is useful for advertisers wishing to reach a larger audience base.
• As used herein, total impressions refers to the total number of collected impressions for particular media regardless of whether multiple ones of those impressions are attributable to the same audience members. That is, multiple impressions accounted for in the total impressions may be attributable to a same audience member.
• Misattribution is a measurement error that typically arises when impressions are collected from a same client device that is shared by multiple people in that a media impression caused by one person that is currently using the client device is incorrectly attributed (i.e., misattributed) to another person that previously used the same client device. Sharing of a client device can occur between two individuals who: (1) live in the same household, and/or (2) have access to the same client device. Misattribution occurs when, for a particular media exposure on a client device, a logged-in-user of a database proprietor service (e.g., Facebook) is not the same as the current user of the client device that is being exposed to the media. For example, if person A logs into the database proprietor's website in the morning on a client device, but person B uses the same client device in the afternoon without logging in (e.g., without user A logging out) and is exposed to media tagged with beacon instructions, the database proprietor attributes the impression to person A since he/she was the last person to log into the database proprietor's site from that client device, while actually it was person B who was using the client device when the media was presented.
• Examples disclosed herein can be used to correct misattribution in collected impressions by applying a misattribution correction to impression data obtained from a database proprietor (e.g., Facebook, Yahoo, Google, etc.) after a profile correction (e.g., a Decision Tree (DT) model) has been applied to the impression data. Examples disclosed herein may be implemented by an audience measurement entity (e.g., any entity interested in measuring or tracking audience exposures to advertisements, content, and/or any other media) in cooperation with any number of database proprietors such as online web services providers. Such database proprietors/online web services providers may be social network sites (e.g., Facebook, Twitter, MySpace, etc.), multi-service sites (e.g., Yahoo!, Google, Axiom, Catalina, etc.), online retailer sites (e.g., Amazon.com, Buy.com, etc.), credit reporting sites (e.g., Experian), and/or any other web service(s) site that maintains user registration records.
• Example methods and/or articles of manufacture comprising computer readable instructions disclosed herein may be used to receive, at a first internet domain, a first request from a client device, the first request indicative of access to media at the client device. In such examples, a response is sent from the first internet domain to the client device. In such examples, the response instructs the client device to send a second request to a second internet domain. In such examples, the second request is to be indicative of the access to the media at the client device. In such examples, an impressions adjustment factor is determined for a first demographic group based on first impressions reported by the client device to the first internet domain and second impressions reported by the client device to the second internet domain. In such example, the first and second impressions correspond to the same media accessed on the client device. In such examples, a misattribution-corrected impressions count is determined for the first demographic group based on the impressions adjustment factor and based on a second impressions count determined at the second internet domain for the first demographic group. In such examples, the second impressions count includes an error based on some of the second impressions being misattributed at the second internet domain to the first demographic group when the some of the second impressions correspond to a second demographic group.
• In some examples, determining the misattribution-corrected impression count involves shifting an impression from the second impressions count corresponding to the first demographic group to a third impressions count corresponding to the second demographic group based on the impressions adjustment factor. In some examples, the first impressions are reported by the client device to an audience measurement entity at the first internet domain that does not provide the media to the client device, and a user of the client device is a panel member of the audience measurement entity. In some examples, the second impressions are reported by the client device to a social network service at the second internet domain to which a user of the client device is subscribed. In some examples, the impressions adjustment factor is to correct impression quantities having inaccuracies due to impressions incorrectly attributed to demographic data not corresponding to persons corresponding to the impressions.
• In some examples, determining the impressions adjustment factor involves subtracting a first unique audience size determined by an audience measurement entity at the first internet domain based on the first impressions from a second unique audience size determined by a database proprietor at the second internet domain based on the second impressions to generate a difference. In such examples, the difference is divided by a total impressions count of the first impressions to determine the impressions adjustment factor.
• In some examples, the error in the second impressions count is based on an entity at the second internet domain incorrectly identifying a user of the client device as belonging to the first demographic group when the user belongs to the second demographic group. In such examples, the misattribution-corrected impressions count comprises fewer impressions than the second impression count based on shifting an impression corresponding to the user from the second impressions count corresponding to the first demographic group to a third impressions count corresponding to the second demographic group based on the impressions adjustment factor.
• In some examples, the misattribution-corrected impressions count is determined based on the impressions adjustment factor without communicating with individual online users about their online media access activities and without using survey responses from the online users to determine the error. In some examples, network communication bandwidth is conserved by not communicating with individual online users about their online media access activities and by not requesting survey responses from the online users to determine the error. In some examples, computer processing resources are conserved by not communicating with individual online users about their online media access activities and by not requesting survey responses from the online users to determine the error.
• Example disclosed apparatus include an example impression collector to receive, at a first internet domain, a first request from a client device, the first request indicative of access to media at the client device. The example impression collector is also to send, from the first internet domain, a response to the client device, the response to instruct the client device to send a second request to a second internet domain, the second request to be indicative of the access to the media at the client device. Such example apparatus also include an impressions adjustment factor determiner to determine an impressions adjustment factor for a first demographic group based on first impressions reported by the client device to the first internet domain and second impressions reported by the client device to the second internet domain. In such examples, the first and second impressions correspond to the same media accessed on the client device. Such example apparatus also includes an impressions corrector to determine a misattribution-corrected impressions count for the first demographic group based on the impressions adjustment factor and based on a second impressions count determined at the second internet domain for the first demographic group. In such examples, the second impressions count includes an error based on some of the second impressions being misattributed at the second internet domain to the first demographic group when the some of the second impressions correspond to a second demographic group.
• In some examples, the impressions corrector is to determine the misattribution-corrected impressions count by shifting an impression from the second impressions count corresponding to the first demographic group to a third impressions count corresponding to the second demographic group based on the impressions adjustment factor. In some examples, the first impressions are reported by the client device to an audience measurement entity at the first internet domain that does not provide the media to the client device. In some examples, a user of the client device is a panel member of the audience measurement entity. In some examples, the second impressions are reported by the client device to a social network service at the second internet domain to which a user of the client device is subscribed.
• In some examples, the impressions adjustment factor determiner is to determine the impressions adjustment factor by subtracting a first unique audience size determined by an audience measurement entity at the first internet domain based on the first impressions from a second unique audience size determined by a database proprietor at the second internet domain based on the second impressions to generate a difference. In such examples, the difference is divided by a total impressions count of the first impressions.
• In some examples, the impressions adjustment factor is to correct impression quantities having inaccuracies due to impressions incorrectly attributed to demographic data not corresponding to persons corresponding to the impressions. In some examples, the error in the second impressions count is based on an entity at the second internet domain incorrectly identifying a user of the client device as belonging to the first demographic group when the user belongs to the second demographic group. In some examples, the misattribution-corrected impressions count include fewer impressions than the second impression count based on shifting an impression corresponding to the user from the second impressions count corresponding to the first demographic group to a third impressions count corresponding to the second demographic group based on the impressions adjustment factor.
• In some examples, the impressions corrector determines the misattribution-corrected impressions based on the impressions adjustment factor without communicating with individual online users about their online media access activities and without using survey responses from the online users to determine the error. In some examples, by determining the misattribution-corrected impressions using the impressions corrector, network communication bandwidth is conserved by not communicating with individual online users about their online media access activities and by not requesting survey responses from the online users to determine the error. In some examples, by determining the misattribution-corrected impressions using the impressions corrector, computer processing resources are conserved by not communicating with individual online users about their online media access activities and by not requesting survey responses from the online users to determine the error.
• Example methods and/or articles of manufacture comprising computer readable instructions disclosed herein may be used to receive, at a first internet domain, a first request from a client device, the first request indicative of access to media at the client device. In such examples, a response is sent from the first internet domain to the client device. In such examples, the response is to instruct the client device to send a second request to a second internet domain. In such examples, the second request is to be indicative of the access to the media at the client device. In such examples, an audience adjustment factor is determined for a demographic group based on first impressions reported by the client device to the first internet domain and second impressions reported by the client device to the second internet domain. In such examples, the first and second impressions correspond to the same media accessed on the client device. In such examples, a misattribution-corrected unique audience size is determined for the demographic group based on the audience adjustment factor and based on a second unique audience size determined at the second internet domain for the demographic group. In such examples, the second unique audience size includes an error based on third impressions misattributed at the second internet domain to the demographic group when the third impressions correspond to another demographic group.
• In some examples, determining the audience adjustment factor involves dividing a third unique audience size corresponding to the first impressions by a fourth unique audience size corresponding to the second impressions. In some examples, determining the misattribution-corrected unique audience size for the demographic group involves dividing the second unique audience size by the audience adjustment factor. In some examples, the first impressions are reported by the client device to an audience measurement entity at the first internet domain that does not provide the media to the client device, and a user of the client device is a panel member of the audience measurement entity. In some examples, the second impressions are reported by the client device to a social network service at the second internet domain to which a user of the client device is subscribed. In some examples, the audience adjustment factor is to correct unique audience size values having inaccuracies due to impressions incorrectly attributed to demographic data not corresponding to persons corresponding to the impressions.
• In some examples, the error in the second unique audience size is based on an entity at the second internet domain incorrectly identifying a user of the client device as belonging to the demographic group when the user belongs to the another demographic group. In some such examples, the misattribution-corrected unique audience size is different than the second unique audience size based on dividing the second unique audience size by the audience adjustment factor.
• In some examples, the misattribution-corrected unique audience size is determined based on the audience adjustment factor without communicating with individual online users about their online media access habits and without using survey responses from the online users to determine the error. In some examples, network communication bandwidth is conserved by not communicating with individual online users about their online media access habits and by not requesting survey responses from the online users to determine the error. In some examples, computer processing resources are conserved by not communicating with individual online users about their online media access habits and by not requesting survey responses from the online users to determine the error.
• Example disclosed apparatus include an example impression collector to receive, at a first internet domain, a first request from a client device. In such examples, the first request is indicative of access to media at the client device. The example impression collector is also to send, from the first internet domain, a response to the client device. In such examples, the response is to instruct the client device to send a second request to a second internet domain. In such examples, the second request is to be indicative of the access to the media at the client device. Such example apparatus also include an audience adjustment factor determiner to determine an audience adjustment factor for a demographic group based on first impressions reported by the client device to the first internet domain and second impressions reported by the client device to the second internet domain. In such examples, the first and second impressions correspond to the same media accessed on the client device. Such example apparatus also include a unique audience corrector to determine a misattribution-corrected unique audience size for the demographic group based on the audience adjustment factor and based on a second unique audience size determined at the second internet domain for the demographic group. In such examples, the second unique audience size includes an error based on third impressions misattributed at the second internet domain to the demographic group when the third impressions correspond to another demographic group.
• In some examples, the audience adjustment factor determiner is to determine the audience adjustment factor by dividing a third unique audience size corresponding to the first impressions by a fourth unique audience size corresponding to the second impressions. In some examples, the unique audience corrector is to determine the misattribution-corrected unique audience size for the demographic group by dividing the second unique audience size by the audience adjustment factor. In some examples, the first impressions are reported by the client device to an audience measurement entity at the first internet domain that does not provide the media to the client device, and a user of the client device is a panel member of the audience measurement entity. In some examples, the second impressions are reported by the client device to a social network service at the second internet domain to which a user of the client device is subscribed. In some examples, the audience adjustment factor is to correct unique audience size values having inaccuracies due to impressions incorrectly attributed to demographic data not corresponding to persons corresponding to the impressions.
• In some examples, the error in the second unique audience size is based on an entity at the second internet domain incorrectly identifying a user of the client device as belonging to the demographic group when the user belongs to the another demographic group. In some such examples, the misattribution-corrected unique audience size comprising dividing the second unique audience size by the audience adjustment factor.
• In some examples, the unique audience corrector is to determine the misattribution-corrected unique audience size based on the audience adjustment factor without communicating with individual online users about their online media access habits and without using survey responses from the online users to determine the error. In some examples, by determining the misattribution-corrected unique audience size, the unique audience corrector conserves network communication bandwidth by not communicating with individual online users about their online media access habits and by not requesting survey responses from the online users to determine the error. In some examples, by determining the misattribution-corrected unique audience size, the unique audience corrector conserves computer processing resources by not communicating with individual online users about their online media access habits and by not requesting survey responses from the online users to determine the error.
• FIG. 1 illustrates anexample client device102 that reports audience impressions for media toimpression collection entities104 to facilitate identifying total impressions and sizes of unique audiences exposed to different media. As used herein, the term impression collection entity refers to any entity that collects impression data. Theclient device102 of the illustrated example may be any device capable of accessing media over a network. For example, theclient device102 may be a computer, a tablet, a mobile device, a smart television, or any other Internet-capable device or appliance. Examples disclosed herein may be used to collect impression information for any type of media including content and/or advertisements. Media may include advertising and/or content such as web pages, streaming video, streaming audio, movies, and/or any other type of content and/or advertisement deliver via satellite, broadcast, cable television, radio frequency (RF) terrestrial broadcast, Internet (e.g., internet protocol television (IPTV)), television broadcasts, radio broadcasts and/or any other vehicle for delivering media. In some examples, media includes user-generated media that is, for example, uploaded to media upload sites such as YouTube and subsequently downloaded and/or streamed by one or more client devices for playback. Media may also include advertisements. Advertisements are typically distributed with content (e.g., programming). Traditionally, content is provided at little or no cost to the audience because it is subsidized by advertisers that pay to have their advertisements distributed with the content. As used herein, “media” refers collectively and/or individually to content and/or advertisement(s) of any type(s).
• In the illustrated example, theclient device102 employs a web browser and/or applications (e.g., apps) to access media, some of which include instructions that cause theclient device102 to report media monitoring information to one or more of theimpression collection entities104. That is, when theclient device102 of the illustrated example accesses media, a web browser and/or application of theclient device102 executes instructions in the media to send a beacon request orimpression request108 to one or more of theimpression collection entities104 via, for example, theInternet110. The beacon requests108 of the illustrated example include information about accesses to media at theclient device102. Such beacon requests108 allow monitoring entities, such as theimpression collection entities104, to collect impressions for different media accessed via theclient device102. In this manner, theimpression collection entities104 can generate large impression quantities for different media (e.g., different content and/or advertisement campaigns).
• Theimpression collection entities104 of the illustrated example include an example audience measurement entity (AME)114 and an example database proprietor (DP)116. In the illustrated example, theAME114 does not provide the media to theclient device102 and is a trusted (e.g., neutral) third party (e.g., The Nielsen Company, LLC) for providing accurate media access statistics. In the illustrated example, thedatabase proprietor116 is one of many database proprietors that operates on the Internet to provide services to large numbers of subscribers. Such services may be email services, social networking services, news media services, cloud storage services, streaming music services, streaming video services, online retail shopping services, credit monitoring services, etc. Example database proprietors include social network sites (e.g., Facebook, Twitter, MySpace, etc.), multi-service sites (e.g., Yahoo!, Google, etc.), online retailer sites (e.g., Amazon.com, Buy.com, etc.), credit reporting services (e.g., Experian) and/or any other web service(s) site that maintains user registration records. In examples disclosed herein, thedatabase proprietor116 maintains user account records corresponding to users registered for Internet-based services provided by the database proprietors. That is, in exchange for the provision of services, subscribers register with thedatabase proprietor116. As part of this registration, the subscribers provide detailed demographic information to thedatabase proprietor116. Demographic information may include, for example, gender, age, ethnicity, income, home location, education level, occupation, etc. In the illustrated example, thedatabase proprietor116 sets a device/user identifier (e.g., an identifier described below in connection withFIG. 2) on a subscriber'sclient device102 that enables thedatabase proprietor116 to identify the subscriber.
• In the illustrated example, when thedatabase proprietor116 receives a beacon/impression request108 from theclient device102, thedatabase proprietor116 requests theclient device102 to provide the device/user identifier that thedatabase proprietor116 had previously set for theclient device102. Thedatabase proprietor116 uses the device/user identifier corresponding to theclient device102 to identify demographic information in its user account records corresponding to the subscriber of theclient device102. In this manner, thedatabase proprietor116 can generate demographic impressions by associating demographic information with an audience impression for the media accessed at theclient device102. As explained above, a demographic impression is an impression that is associated with a characteristic (e.g., a demographic characteristic) of the person exposed to the media.
• In the illustrated example, theAME114 establishes an AME panel of users who have agreed to provide their demographic information and to have their Internet browsing activities monitored. When an individual joins the AME panel, the person provides detailed information concerning the person's identity and demographics (e.g., gender, age, ethnicity, income, home location, occupation, etc.) to theAME114. TheAME114 sets a device/user identifier (e.g., an identifier described below in connection withFIG. 2) on the person'sclient device102 that enables theAME114 to identify the panelist. An AME panel may be a cross-platform home television/computer (TVPC) panel built and maintained by theAME114. In other examples, the AME panel may be a computer panel or internet-device panel without corresponding to a television audience panel. In yet other examples, the AME panel may be a cross-platform radio/computer panel and/or a panel formed for other mediums.
• In the illustrated example, when theAME114 receives abeacon request108 from theclient device102, theAME114 requests theclient device102 to provide theAME114 with the device/user identifier that theAME114 previously set in theclient device102. TheAME114 uses the device/user identifier corresponding to theclient device102 to identify demographic information in its user AME panelist records corresponding to the panelist of theclient device102. In this manner, theAME114 can generate demographic impressions by associating demographic information with an audience impression for the media accessed at theclient device102.
• In the illustrated example, theclient device102 is used in anexample household120 in whichhousehold members122 and124 (identified assubscriber A122 and subscriber B124) are subscribers of an internet-based service offered by thedatabase proprietor116. In the illustrated example,subscriber A122 andsubscriber B124 share theclient device102 to access the internet-based service of thedatabase proprietor116 and to access other media via theInternet110. In the illustrated example, when thedatabase proprietor116 receives a beacon/impression request108 for media accessed via theclient device102, thedatabase proprietor116 logs an impression for the media access as corresponding to thesubscriber122,124 of thehousehold120 that most recently logged into thedatabase proprietor116. Misattributions of impressions logged by thedatabase proprietor116 are likely to occur in circumstances similar to theexample household120 ofFIG. 1 in which multiple people in a household share a client device. For example, if thesubscriber A122 logs into a service of thedatabase proprietor116 on theclient device102, and thesubscriber B124 subsequently uses theclient device102 without logging in to the service of thedatabase proprietor116, thedatabase proprietor116 attributes logged impression to thesubscriber A122 even though the use is actually bysubscriber B124 because thesubscriber A122 was the last person to log into thedatabase proprietor116 and, thus, thesubscriber A122 was most recently identified by thedatabase proprietor116 as the subscriber using theclient device102. As such, even though thesubscriber B124 was subsequently using theclient device102, impressions logged by thedatabase proprietor116 during such use are not attributed to the correct person (i.e., the subscriber B124) because the most recent login detected by thedatabase proprietor116 corresponded to thesubscriber A122. In the illustrated example, logins are used by thedatabase proprietor116 to identify subscribers using particular devices by associating device/user identifiers on the client devices with subscriber accounts at thedatabase proprietor116 corresponding to usernames used during the logins. As such, thedatabase proprietor116 assumes that the most recent login is indicative of a subscriber using theclient device102 until another login event is received at thedatabase proprietor116 that identifies a different subscriber. Such assumptions based on the most recent login lead to the above-described misattributions.
• FIG. 2 illustrates an example communication flow diagram of an example manner in which theAME114 and thedatabase proprietor116 ofFIG. 1 can collect impressions and demographic information based on theclient device102 reporting impressions to theAME114 and thedatabase proprietor116.FIG. 2 also shows anexample misattribution corrector202. Themisattribution corrector202 of the illustrated example is to correct unique audience sizes and impression counts that are based on impressions reported by client devices (e.g., the client device102) and for which thedatabase proprietor116 has misattributed some of those impressions to incorrect people and, thus, incorrect demographic information. The example chain of events shown inFIG. 2 occurs when theclient device102 accesses media for which theclient device102 reports an impression to theAME114 and thedatabase proprietor116. In some examples, theclient device102 reports impressions for accessed media based on instructions (e.g., beacon instructions) embedded in the media that instruct the client device102 (e.g., instruct a web browser or an app in the client device102) to send beacon/impression requests (e.g., the beacon/impression requests108 ofFIG. 1) to theAME114 and/or thedatabase proprietor116. In such examples, the media having the beacon instructions is referred to as tagged media. In other examples, theclient device102 reports impressions for accessed media based on instructions embedded in apps or web browsers that execute on theclient device102 to send beacon/impression requests (e.g., the beacon/impression requests108 ofFIG. 1) to theAME114, and/or thedatabase proprietor116 for corresponding media accessed via those apps or web browsers. In any case, the beacon/impression requests (e.g., the beacon/impression requests108 ofFIG. 1) include device/user identifiers (e.g., AME IDs and/or DP IDs) as described further below to allow thecorresponding AME114 and/ordatabase proprietor116 to associate demographic information with resulting logged impressions.
• In the illustrated example, theclient device102accesses media206 that is tagged withbeacon instructions208. Thebeacon instructions208 cause theclient device102 to send a beacon/impression request212 to anAME impressions collector218 when theclient device102 accesses themedia206. For example, a web browser and/or app of theclient device102 executes thebeacon instructions208 in themedia206 which instruct the browser and/or app to generate and send the beacon/impression request212. In the illustrated example, theclient device102 sends the beacon/impression request212 to theAME impression collector218 using an HTTP (hypertext transfer protocol) request addressed to the URL (uniform resource locator) of theAME impressions collector218 at, for example, a first internet domain of theAME114. The beacon/impression request212 of the illustrated example includes a media identifier213 (e.g., an identifier that can be used to identify content, an advertisement, and/or any other media) corresponding to themedia206. In some examples, the beacon/impression request212 also includes a site identifier (e.g., a URL) of the website that served themedia206 to theclient device102 and/or a host website ID (e.g., www.acme.com) of the website that displays or presents themedia206. In the illustrated example, the beacon/impression request212 includes a device/user identifier214. In the illustrated example, the device/user identifier214 that theclient device102 provides in thebeacon impression request212 is an AME ID because it corresponds to an identifier that theAME114 uses to identify a panelist corresponding to theclient device102. In other examples, theclient device102 may not send the device/user identifier214 until theclient device102 receives a request for the same from a server of the AME114 (e.g., in response to, for example, theAME impressions collector218 receiving the beacon/impression request212).
• In some examples, the device/user identifier214 may be a device identifier (e.g., an international mobile equipment identity (IMEI), a mobile equipment identifier (MEID), a media access control (MAC) address, etc.), a web browser unique identifier (e.g., a cookie), a user identifier (e.g., a user name, a login ID, etc.), an Adobe Flash client identifier, identification information stored in an HTML5 datastore, and/or any other identifier that theAME114 stores in association with demographic information about users of theclient devices102. When theAME114 receives the device/user identifier214, theAME114 can obtain demographic information corresponding to a user of theclient device102 based on the device/user identifier214 that theAME114 receives from theclient device102. In some examples, the device/user identifier214 may be encrypted (e.g., hashed) at theclient device102 so that only an intended final recipient of the device/user identifier214 can decrypt the hashedidentifier214. For example, if the device/user identifier214 is a cookie that is set in theclient device102 by theAME114, the device/user identifier214 can be hashed so that only theAME114 can decrypt the device/user identifier214. If the device/user identifier214 is an IMEI number, theclient device102 can hash the device/user identifier214 so that only a wireless carrier (e.g., the database proprietor116) can decrypt the hashedidentifier214 to recover the IMEI for use in accessing demographic information corresponding to the user of theclient device102. By hashing the device/user identifier214, an intermediate party (e.g., an intermediate server or entity on the Internet) receiving the beacon request cannot directly identify a user of theclient device102.
• In response to receiving the beacon/impression request212, theAME impressions collector218 logs an impression for themedia206 by storing themedia identifier213 contained in the beacon/impression request212. In the illustrated example ofFIG. 2, theAME impressions collector218 also uses the device/user identifier214 in the beacon/impression request212 to identify AME panelist demographic information corresponding to a panelist of theclient device102. That is, the device/user identifier214 matches a user ID of a panelist member (e.g., a panelist corresponding to a panelist profile maintained and/or stored by the AME114). In this manner, theAME impressions collector218 can associate the logged impression with demographic information of a panelist corresponding to theclient device102.
• In some examples, the beacon/impression request212 may not include the device/user identifier214 if, for example, the user of theclient device102 is not an AME panelist. In such examples, theAME impressions collector218 logs impressions regardless of whether theclient device102 provides the device/user identifier214 in the beacon/impression request212 (or in response to a request for the identifier214). When theclient device102 does not provide the device/user identifier214, theAME impressions collector218 will still benefit from logging an impression for themedia206 even though it will not have corresponding demographics. For example, theAME114 may still use the logged impression to generate a total impressions count and/or a frequency of impressions (e.g., an impressions frequency) for themedia206. Additionally or alternatively, theAME114 may obtain demographics information from thedatabase proprietor116 for the logged impression if theclient device102 corresponds to a subscriber of thedatabase proprietor116.
• In the illustrated example ofFIG. 2, to compare or supplement panelist demographics (e.g., for accuracy or completeness) of theAME114 with demographics from one or more database proprietors (e.g., the database proprietor116), theAME impressions collector218 returns a beacon response message222 (e.g., a first beacon response) to theclient device102 including an HTTP “302 Found” re-direct message and a URL of a participatingdatabase proprietor116 at, for example, a second internet domain. In the illustrated example, the HTTP “302 Found” re-direct message in thebeacon response222 instructs theclient device102 to send asecond beacon request226 to thedatabase proprietor116. In other examples, instead of using an HTTP “302 Found” re-direct message, redirects may be implemented using, for example, an iframe source instruction (e.g., <iframe src=“ ”>) or any other instruction that can instruct a client device to send a subsequent beacon request (e.g., the second beacon request226) to a participatingdatabase proprietor116. In the illustrated example, theAME impressions collector218 determines thedatabase proprietor116 specified in thebeacon response222 using a rule and/or any other suitable type of selection criteria or process. In some examples, theAME impressions collector218 determines a particular database proprietor to which to redirect a beacon request based on, for example, empirical data indicative of which database proprietor is most likely to have demographic data for a user corresponding to the device/user identifier214. In some examples, thebeacon instructions208 include a predefined URL of one or more database proprietors to which theclient device102 should send follow up beacon requests226. In other examples, the same database proprietor is always identified in the first redirect message (e.g., the beacon response222).
• In the illustrated example ofFIG. 2, the beacon/impression request226 may include a device/user identifier227 that is a DP ID because it is used by thedatabase proprietor116 to identify a subscriber of theclient device102 when logging an impression. In some instances (e.g., in which thedatabase proprietor116 has not yet set a DP ID in the client device102), the beacon/impression request226 does not include the device/user identifier227. In some examples, the DP ID is not sent until the DP requests the same (e.g., in response to the beacon/impression request226). In some examples, the device/user identifier227 is a device identifier (e.g., an international mobile equipment identity (IMEI), a mobile equipment identifier (MEID), a media access control (MAC) address, etc.), a web browser unique identifier (e.g., a cookie), a user identifier (e.g., a user name, a login ID, etc.), an Adobe Flash client identifier, identification information stored in an HTML5 datastore, and/or any other identifier that thedatabase proprietor116 stores in association with demographic information about subscribers corresponding to theclient devices102. When thedatabase proprietor116 receives the device/user identifier227, thedatabase proprietor116 can obtain demographic information corresponding to a user of theclient device102 based on the device/user identifier227 that thedatabase proprietor116 receives from theclient device102. In some examples, the device/user identifier227 may be encrypted (e.g., hashed) at theclient device102 so that only an intended final recipient of the device/user identifier227 can decrypt the hashedidentifier227. For example, if the device/user identifier227 is a cookie that is set in theclient device102 by thedatabase proprietor116, the device/user identifier227 can be hashed so that only thedatabase proprietor116 can decrypt the device/user identifier227. If the device/user identifier227 is an IMEI number, theclient device102 can hash the device/user identifier227 so that only a wireless carrier (e.g., the database proprietor116) can decrypt the hashedidentifier227 to recover the IMEI for use in accessing demographic information corresponding to the user of theclient device102. By hashing the device/user identifier227, an intermediate party (e.g., an intermediate server or entity on the Internet) receiving the beacon request cannot directly identify a user of theclient device102. For example, if the intended final recipient of the device/user identifier227 is thedatabase proprietor116, theAME114 cannot recover identifier information when the device/user identifier227 is hashed by theclient device102 for decrypting only by the intendeddatabase proprietor116.
• In some examples that use cookies as the device/user identifier227, when a user deletes a database proprietor cookie from theclient device102, thedatabase proprietor116 sets the same cookie value in theclient device102 the next time the user logs into a service of thedatabase proprietor116. In such examples, the cookies used by thedatabase proprietor116 are registration-based cookies, which facilitate setting the same cookie value after a deletion of the cookie value has occurred on theclient device102. In this manner, thedatabase proprietor116 can collect impressions for theclient device102 based on the same cookie value over time to generate unique audience (UA) sizes while eliminating or substantially reducing the likelihood that a single unique person will be counted as two or more separate unique audience members.
• Although only asingle database proprietor116 is shown inFIGS. 1 and 2, the impression reporting/collection process ofFIGS. 1 and 2 may be implemented using multiple database proprietors. In some such examples, thebeacon instructions208 cause theclient device102 to send beacon/impression requests226 to numerous database proprietors. For example, thebeacon instructions208 may cause theclient device102 to send the beacon/impression requests226 to the numerous database proprietors in parallel or in daisy chain fashion. In some such examples, thebeacon instructions208 cause theclient device102 to stop sending beacon/impression requests226 to database proprietors once a database proprietor has recognized theclient device102. In other examples, thebeacon instructions208 cause theclient device102 to send beacon/impression requests226 to database proprietors so that multiple database proprietors can recognize theclient device102 and log a corresponding impression. In any case, multiple database proprietors are provided the opportunity to log impressions and provide corresponding demographics information if the user of theclient device102 is a subscriber of services of those database proprietors.
• In some examples, prior to sending thebeacon response222 to theclient device102, theAME impressions collector218 replaces site IDs (e.g., URLs) of media provider(s) that served themedia206 with modified site IDs (e.g., substitute site IDs) which are discernible only by theAME114 to identify the media provider(s). In some examples, theAME impressions collector218 may also replace a host website ID (e.g., www.acme.com) with a modified host site ID (e.g., a substitute host site ID) which is discernible only by theAME114 as corresponding to the host website via which themedia206 is presented. In some examples, theAME impressions collector218 also replaces themedia identifier213 with a modifiedmedia identifier213 corresponding to themedia206. In this way, the media provider of themedia206, the host website that presents themedia206, and/or themedia identifier213 are obscured from thedatabase proprietor116, but thedatabase proprietor116 can still log impressions based on the modified values which can later be deciphered by theAME114 after theAME114 receives logged impressions from thedatabase proprietor116. In some examples, theAME impressions collector218 does not send site IDs, host site IDS, themedia identifier213 or modified versions thereof in thebeacon response222. In such examples, theclient device102 provides the original, non-modified versions of themedia identifier213, site IDs, host IDs, etc. to thedatabase proprietor116.
• In the illustrated example, theAME impression collector218 maintains a modified ID mapping table228 that maps original site IDs with modified (or substitute) site IDs, original host site IDs with modified host site IDs, and/or maps modified media identifiers to the media identifiers such as themedia identifier213 to obfuscate or hide such information from database proprietors such as thedatabase proprietor116. Also in the illustrated example, theAME impressions collector218 encrypts all of the information received in the beacon/impression request212 and the modified information to prevent any intercepting parties from decoding the information. TheAME impressions collector218 of the illustrated example sends the encrypted information in thebeacon response222 to theclient device102 so that theclient device102 can send the encrypted information to thedatabase proprietor116 in the beacon/impression request226. In the illustrated example, theAME impressions collector218 uses an encryption that can be decrypted by thedatabase proprietor116 site specified in the HTTP “302 Found” re-direct message.
• Periodically or aperiodically, the impression data collected by thedatabase proprietor116 is provided to aDP impressions collector230 of theAME114 as, for example, batch data. As discussed above, some impressions logged by theclient device102 to thedatabase proprietor116 are misattributed by thedatabase proprietor116 to a wrong subscriber and, thus, to incorrect demographic information. During a data collecting and merging process to combine demographic and impression data from theAME114 and thedatabase proprietor116, demographics of impressions logged by theAME114 for theclient device102 will not correspond to demographics of impressions logged by thedatabase proprietor116 because thedatabase proprietor116 has misattributed some impressions to the incorrect demographic information. Examples disclosed herein may be used to determine an impressions adjustment factor to correct/adjust impression-based data (e.g., total impressions and unique audience size) provided by thedatabase proprietor116.
• Additional examples that may be used to implement the beacon instruction processes ofFIG. 2 are disclosed in Mainak et al., U.S. Pat. No. 8,370,489, which is hereby incorporated herein by reference in its entirety. In addition, other examples that may be used to implement such beacon instructions are disclosed in Blumenau, U.S. Pat. No. 6,108,637, which is hereby incorporated herein by reference in its entirety.
• In the example ofFIG. 2, theAME114 includes theexample misattribution corrector202 to correct unique audience values and impression counts that are based on impressions reported by client devices (e.g., the client device102) for which thedatabase proprietor116 has misattributed some of the impressions to incorrect demographic information. Themisattribution corrector202 of the illustrated example is provided with an example audienceadjustment factor determiner232, an example impressionsadjustment factor determiner234, an exampleunique audience corrector236, and anexample impressions corrector238.
• The example audienceadjustment factor determiner232 ofFIG. 2 is provided to calculate a unique audience (UA) adjustment factor representative of an inaccurate UA size that is based on misattributed impressions relative to a UA size that is based on accurately attributed impressions. As discussed above, misattribution occurs when thedatabase proprietor116 identifies the wrong person as being a current user of theclient device102 when the client device reports an impression for accessed media to thedatabase proprietor116. The example impressionsadjustment factor determiner234 is provided to calculate an impressions adjustment factor representative of an amount of misattributed impressions relative to an amount of correctly attributed impressions.
• The exampleunique audience corrector236 ofFIG. 2 is provided to correct unique audience sizes or quantities by applying the impressions adjustment factor (determined by the impressions adjustment factor determiner234) to total unique audience sizes corresponding to total impressions collected by theAME114. The example impressions corrector238 is provided to correct an impressions count by applying the impressions adjustment factor (determined by the impressions adjustment factor determiner234) to the total number of impressions collected by theAME114.
• Although themisattribution corrector202 is shown in the illustrated example as being located in theAME114, themisattribution corrector202 may alternatively be located at any other location such as at thedatabase proprietor116 or at any other suitable location (e.g., location(s) separate from theAME114 and the database proprietor116). In addition, although theAME impressions collector218, the modifiedID map228, and theDP impressions collector230 are shown separate from themisattribution corrector202, one or more of theAME impressions collector218, the modifiedID map228, and/or theDP impressions collector230 may be implemented in themisattribution corrector202.
• While an example manner of implementing theexample misattribution corrector202, theexample impressions collector218, the example modifiedID map228, the exampleDP impressions collector230, the example audienceadjustment factor determiner232, the example impressionsadjustment factor determiner234, the exampleunique audience corrector236, and the example impressions corrector238 is illustrated inFIG. 2, one or more of the elements, processes and/or devices illustrated inFIG. 2 may be combined, divided, re-arranged, omitted, eliminated and/or implemented in any other way. Further, theexample misattribution corrector202, the exampleAME impressions collector218, the example modifiedID map228, the exampleDP impressions collector230, the example audienceadjustment factor determiner232, the example impressionsadjustment factor determiner234, the exampleunique audience corrector236, and/or the example impressions corrector238 ofFIG. 2 may be implemented by hardware, software, firmware and/or any combination of hardware, software, and/or firmware. Thus, for example, any of theexample misattribution corrector202, the exampleAME impressions collector218, the example modifiedID map228, the exampleDP impressions collector230, the example audienceadjustment factor determiner232, the example impressionsadjustment factor determiner234, the exampleunique audience corrector236, and/or the example impressions corrector238 could be implemented by one or more analog or digital circuit(s), logic circuits, programmable processor(s), application specific integrated circuit(s) (ASIC(s)), programmable logic device(s) (PLD(s)), and/or field programmable logic device(s) (FPLD(s)). When reading any of the apparatus or system claims of this patent to cover a purely software and/or firmware implementation, at least one of theexample misattribution corrector202, the exampleAME impressions collector218, the example modifiedID map228, the exampleDP impressions collector230, the example audienceadjustment factor determiner232, the example impressionsadjustment factor determiner234, the exampleunique audience corrector236, and/or the example impressions corrector238 is/are hereby expressly defined to include a tangible computer readable storage device or storage disk such as a memory, a digital versatile disk (DVD), a compact disk (CD), a Blu-ray disk, etc. storing the software and/or firmware. Further still, theexample misattribution corrector202, theexample impressions collector218, the example modifiedID map228, the exampleDP impressions collector230, the example audienceadjustment factor determiner232, the example impressionsadjustment factor determiner234, the exampleunique audience corrector236, and/or the example impressions corrector238 ofFIG. 2 may include one or more elements, processes and/or devices in addition to, or instead of, those illustrated inFIG. 2, and/or may include more than one of any or all of the illustrated elements, processes and devices.
• Examples disclosed herein to correct impression-based data (e.g., total impressions and unique audience size) provided by thedatabase proprietor116 involve generating an adjustment factor based on impressions collected by theAME116 and correctly attributed to demographic information for corresponding AME panelists. Themisattribution corrector202 ofFIG. 2 may be implemented using the example techniques below to correct impression-based data that is based on impressions of which some are misattributed to the wrong demographic information by thedatabase proprietor116.
• Examples disclosed herein involve using impressions logged by theAME114 in association with demographic data collected from AME panel members to calculate an audience adjustment factor usingexample Equation 1 below and an impression adjustment factor usingexample Equation 2 below. Audience adjustment factors determined usingexample Equation 1 can be used to correct unique audience size values having inaccuracies due to misattributions of impressions by database proprietors. Impression adjustment factors determined usingexample Equation 2 below can be used to correct impression quantities having inaccuracies due to misattributions of impressions by database proprietors.
• In the illustrated example ofFIG. 2, the audienceadjustment factor determiner232 can useexample Equation 1 below to determine an audience adjustment factor (f_i,j) for persons in a demographic group (j) that accessed media (i).
• $\begin{array}{cc}{f}_{i,j}=\frac{\sum _{i,j}^{}F_{i,j}}{\sum _{i,j}^{}T_{i,j}}& \mathrm{Equation}1\end{array}$
• Inexample Equation 1 above, f_i,jis the adjustment factor for a unique audience (UA) size of a particular demographic group (j) that accessed media (i), F_i,jis a database proprietor (DP) UA count of the number of AME panelists of theAME114 that thedatabase proprietor116 observes (e.g., recognizes, identifies, logs impressions for, etc.) in the demographic group (j) as accessing the media (i), and T_i,jis an AME UA count of AME panelists that theAME114 observes in the demographic group (j) as accessing the media (i).
• In the illustrated example ofFIG. 2, the impressionsadjustment factor determiner234 employsexample Equation 2 below to determine an impressions adjustment factor (k_i,j) for persons in a demographic group (j) that accessed media (i).
• $\begin{array}{cc}\begin{array}{c}{k}_{i,j}=\frac{\sum _{i,j}^{}Q_{i,j}}{\sum _i^{}S_i}-\frac{\sum _{i,j}^{}R_{i,j}}{\sum _i^{}S_i}\\ =\frac{\sum _{i,j}^{}Q_{i,j}-\sum _{i,j}^{}R_{i,j}}{\sum _i^{}S_i}\end{array}& \mathrm{Equation}2\end{array}$
• Inexample Equation 2 above, k_i,jis the impressions adjustment factor for impressions logged for a particular demographic group (j) that accessed media (i), R_i,jis a DP UA count of the number of AME panelists of theAME114 that thedatabase proprietor116 observes (e.g., recognizes, identifies, logs impressions for, etc.) in the demographic group (j) as accessing the media (i), Q_i,jis an AME UA count of AME panelists that theAME114 observes in the demographic group (j) as accessing the media (i), and S_iis the total AME impressions of AME panelists (summed across all demographic groups) that accessed media (i).
• FIG. 3 illustrates an example table300 withexample AME impressions302 collected by theAME114 andexample DP impressions304 collected by thedatabase proprietor116 for different demographic groups (e.g., females younger than 50 years (F<50),females 50 years old and older (F>=50), males younger than 50 years (M<50), andmales 50 years old and older (M>=50)). Theexample AME impressions304 and theexample DP impressions304 shown in the example table300 are development or test impressions that are collected by theAME114 and theDP116 during an adjustment factors development phase (e.g., an adjustment factorsdevelopment phase802 ofFIG. 8) with the purpose of calculating adjustment factors (e.g., audience adjustment (AA) factors402 ofFIGS. 4 and 6 and impression adjustment (IA)factors502 ofFIGS. 5 and 6) that can be subsequently used on large logs of real impressions collected by thedatabase proprietor116 to correct for impression misattributions that affect unique audience sizes and impression counts that are generated using the database proprietor's logged impressions.
• In the illustrated example ofFIG. 3, theDP impressions304 have anexample misattribution error308 for impression #9 (IMP #9). That is, theimpressions302,304 collected by both theAME114 and thedatabase proprietor116 are based on client devices (e.g., the client device102) having users that are both (1) AME panelists of theAME114 and (2) registered subscribers of thedatabase proprietor116. When theAME114 logs an impression based on, for example, the beacon/impression request212 ofFIG. 2 from a panelist of theAME114, theAME114 logs an accurate impression. In the illustrated example ofFIG. 3,such AME impressions302 are also referred to astruth impressions302 because theAME114 regards them as correctly associated with corresponding demographic information of the current user of theclient device102. In some examples, to assure the accuracy of theAME impressions302, theAME114 incentivizes (e.g., through cash or other rewards) AME panel members to login to an AME website whenever the AME panel members begin using aclient device102. In this manner, theAME114 can accurately set and/or associate an AME ID (e.g., the device/user identifier214 ofFIG. 2) with an AME panelist that is currently using theclient device102.
• Unlike the known accuracy, or truth, of theAME impressions302, there are no assurances that theDP impressions304 are accurately associated with correct demographic information. That is, subscribers of thedatabase proprietor116 may not be incentivized to login to a website or service of thedatabase proprietor116 when the subscribers begin using aclient device102. As such, thedatabase proprietor116 is sometimes unable to accurately set and/or associate a DP ID (e.g., the device/user identifier227 ofFIG. 2) with a person that is currently using theclient device102. The misattributions present in the development or test impressions of the table300 ofFIG. 3 are representative of the types of misattributions that thedatabase proprietor116 is likely to make when logging impressions for persons that may or may not be AME panelists and/or may or may not be known to thedatabase proprietor116. Therefore, calculating adjustment factors based on the development impressions of the table300 ofFIG. 3 results in adjustment factors that can be used to correct for misattributions in impressions subsequently collected by thedatabase proprietor116 for other users.
• In the illustrated example ofFIG. 3, theexample misattribution error308 atimpression #9 is created when thedatabase proprietor116 mis-recognizes an impression reported by the client device102 (e.g., via the beacon/impression request226 ofFIG. 2) as being associated with demographic information for a male (M) ofage 30. In the example table300 ofFIG. 3, a correct demographic impression atIMP #9 logged by theAME114 for the same person (e.g., via the beacon/impression request212 ofFIG. 2) shows that the correct demographics indicate that the actual person corresponding to the impression is a female (F) ofage 29.Example Equations 1 and 2 above may be used to correct unique audience sizes and total impression counts that are affected by misattribution errors such as themisattribution error308 ofFIG. 3.
• FIG. 4 illustrates an example table400 with example audience adjustment (AA) factors402 (e.g., the audience adjustment factor (f_i,j) ofEquation 1 above) for unique audience sizes of different demographic (DEMO) groups. Based on theAME impressions302 ofFIG. 3, the audienceadjustment factor determiner232 ofFIG. 2 can useEquation 1 above to calculate the example audience adjustment factors402 for different demographic groups (j) that access particular media (i). The example table400 showsAME UA sizes404 and exampleDP UA sizes406 for different demographic groups. The exampleAME UA sizes404 correspond to the term (T_i,j) ofEquation 1 above, and theDP UA sizes406 correspond to the term (F_i,j) ofEquation 1 above. In the illustrated example ofFIG. 4, theAME UA sizes404 show that theAME impressions302 ofFIG. 3 include three (3) unique audience members of the F<50 demo group, one (1) unique audience member of the F>=50 demo group, two (2) unique audience members of the M<50 demo group, and one (1) unique audience member of the M>=50 demo group. The exampleDP UA sizes406 show that theDP impressions304 ofFIG. 3 include two (2) unique audience members of the F<50 demo group, one (1) unique audience member of the F>=50 demo group, two (2) unique audience members of the M<50 demo group, and one (1) unique audience member of the M>=50 demo group.
• The exampleDP UA sizes406 have a misattribution-basederror410 for the F<50 demo group which results from themisattribution error308 ofFIG. 3. That is, themisattribution error308 ofFIG. 3 mistakenly identifies impression9 (IMP #9) as corresponding to a male (M) ofage 30 rather than the correct demographic of female (F) ofage 29, as noted in theAME impression302. Because of themisattribution error308 ofFIG. 3, impression9 (IMP #9) for theDP impressions304 is not counted for a female (F) ofage 29. Therefore, theDP UA406 ofFIG. 4 for the F<50 demo group is only two (2), which is less than the correct (truth) unique audience for the F<50 demo group of three (3) as shown by the correspondingAME UA size404. Because there were no other misattribution errors in the example impressions ofFIG. 3, theDP UA sizes406 match correspondingAME UA sizes404 for the other demo groups.
• In the illustrated example, the audienceadjustment factor determiner232 ofFIG. 2 usesEquation 1 above to determine the AA factors402. For example, for each of the demo groups F<50, F>=50, M<50, and M>=50, the audienceadjustment factor determiner232 divides the corresponding AME UA404 (T_i,j) by the corresponding DP UA406 (F_i,j) to determine thecorresponding AA factor402 for that demo group. As shown in the example table400, theAA factor402 corresponding to theDP UA406 having the misattribution-basederror410, the correspondingAA factor402 is 0.67 (e.g., T_i,j/F_i,j=>3/2=0.67).
• FIG. 5 illustrates an example table500 with example impression adjustment (IA) factors502 (e.g., the impressions adjustment factor (k_i,j) ofEquation 2 above) for total AME impression counts504 and total DP impression counts506 of different demographic groups (j) determined based on theexample AME impressions302 and theexample DP impressions304 ofFIG. 3. In the illustrated example ofFIG. 5, a misattribution-basederror508 occurs in association with the F<50 demographic group, and a misattribution-basederror510 occurs in association with the M<50 demographic group. The misattribution-basederrors508,510 occur because of themisattribution error308 ofFIG. 3. That is, since themisattribution error308 incorrectly indicates a male (M) ofage 30 instead of the correct female (F) ofage 29, the DP impressions count506 for the F<50 demographic group has one fewer impression than the correct (truth) AME impressions count504 for the F<50 demographic group. In addition, the DP impressions count506 for the M<50 demographic group has one more impression than the correct (truth) AME impressions count for the M<50 demographic group.
• In the illustrated example, the impressionsadjustment factor determiner234 ofFIG. 2 usesEquation 2 above to determine the IA factors502. For example, for each of the demo groups F<50, F>=50, M<50, and M>=50, the impressionsadjustment factor determiner234processes Equation 2 using the corresponding AME impressions504 (Q_i,j) ofFIG. 5, the corresponding DP impressions506 (R_i,j) ofFIG. 5, and the total AME impressions count (S_i) summed across all demographic groups, to determine the correspondingIA factor502 for that demo group. For example, usingEquation 2 above, the impressionsadjustment factor determiner234 subtracts the DP impressions506 (R_i,j) of a particular demographic group from the AME impressions504 (Q_i,j) of the same demographic group, and divides the resulting difference by the total AME impressions count (S_i) summed across all demographic groups (e.g.,IA factor502=((AME impressions504 (Q_i,j))−(DP impressions506 (R_i,j)))/(total AME impressions count (S_i))).
• As shown in the example table500, theIA factor502 corresponding to the F<50 demographic group having the misattribution-basederror508 is 11.11%, and theIA factor502 corresponding to the M<50 demographic group having the misattribution-basederror510 is −11.11%. In the illustrated example, the IA factors502 are 0.0% for the demographic groups not having misattribution-based errors. In the illustrated example, the impressionsadjustment factor determiner234 determines the misattribution-basederror508 of 11.11% for the F<50 demographic group based onEquation 2 above by subtracting the DP impressions506 (R_i,j) of 3 for the F<50 demographic group shown inFIG. 5 from the AME impressions504 (Q_i,j) of 4 for the F<50 demographic group shown inFIG. 5, and divides the resulting difference of one (1) by the total AME impressions count (S_i) of nine (9). Also in the illustrated example, the impressionsadjustment factor determiner234 determines the misattribution-basederror508 of −11.11% for the M<50 demographic group based onEquation 2 above by subtracting the DP impressions506 (R_i,j) of 4 for the M<50 demographic group shown inFIG. 5 from the AME impressions504 (Q_i,j) of 3 for the M<50 demographic group shown inFIG. 5, and divides the resulting difference of negative one (−1) by the total AME impressions count (S_i) of nine (9).
• The IA factors502 of the illustrated example are percentages of the total AME impressions count504 summed across all demographic groups. Thus, theIA factor502 of 11.11% corresponding to the F<50 demographic group means that 11.11% of 9 total AME impressions (S_i) (i.e., the sum of all of theAME impressions504 logged across all of the demographic groups shown inFIG. 5) need to be added to the DP impressions count506 for the F<50 demographic group. For example, 11.11% of nine (9) total AME impressions is one (1), which can be added to the three (3) DP impressions count506 for the F<50 demographic group to make the DP impressions count506 equal to the AME impressions count504 for the F<50 demographic group. In addition, theIA factor502 of −11.11% corresponding to the M<50 demographic group means that −11.11% of the nine (9) total AME impressions (i.e., the sum of all of theAME impressions504 logged across all of the demographic groups shown inFIG. 5) need to be added (or 11.11% need to be subtracted) from the DP impressions count506 for the M<50 demographic group. For example, −11.11% of nine (9) total AME impressions is negative one (−1), which can be added to the four (4) DP impressions count506 for the M<50 demographic group to make the DP impressions count506 equal to the AME impressions count504 for the M<50 demographic group. Thus, the effect of the 11.11% IA factor502 for the F<50 demographic group and the −11.11% IA factor502 for the M<50 demographic group is that one (1)DP impression506 is shifted away from the M<50 demographic group to the F<50 demographic group. In this manner, thetotal DP impressions506 summed across all of the demographic groups remains the same after applying the IA factors502.
• FIG. 6 illustrates an example table600 with misattribution-corrected UA size values602 and misattribution-corrected impression counts604 based on the AA factors402 ofFIG. 4 and the IA factors502 ofFIG. 5 for different demographic groups. The data of example table600 illustrates how UA size values and impression counts received by theAME114 in the aggregate (e.g., not individual impression records) from thedatabase proprietor116 can be adjusted to correct for misattribution-based errors. The aggregate DP UA size values are shown in the example table600 as DP decision tree (DT)-corrected UA size values606. The aggregate DP impression count values are shown in the example table600 as DP DT-corrected impression counts608. To generate the DP DT-corrected UA size values606 and the DP DT-corrected impression counts608, thedatabase proprietor116 performs a profile correction by applying a DT model on demographic data used to log impressions. That is, during initial registration with thedatabase subscriber116, some subscribers may provide inaccurate demographic information and/or may omit certain demographic information. To fill in some of the missing demographic information in subscriber accounts, thedatabase proprietor116 processes the demographic data in subscriber accounts using a DT model that produces the most likely outcomes for the missing demographic data. Any suitable DT model can be used by thedatabase proprietor116 to correct profile data for subscribers of thedatabase proprietor116.
• In the illustrated example ofFIG. 6, the exampleunique audience corrector236 ofFIG. 2 applies the AA factors402 to the DT-corrected UA size values606 to determine the misattribution-corrected UA size values602. That is, theunique audience corrector236 divides a DT-correctedUA size value606 for a demographic group by acorresponding AA factor402 for the same demographic group to calculate a corresponding misattribution-corrected UA size value602 (e.g., (misattribution-corrected UA size)=(DT-corrected UA size)/(AA factor)). For example, for the F<50 demographic group, theunique audience corrector236 divides the DT-correctedUA size value606 of 63,000 by the correspondingAA factor402 of 0.67 to calculate the misattribution-correctedUA size value602 of 94,500 (e.g., 94,500=63,000/0.67). Thus, using the AA factors402 in this manner to calculate the misattribution-corrected UA size values602 substantially reduces or eliminates the effects that misattributed impressions logged by thedatabase proprietor116 have on the DT-corrected UA size values606.
• In the illustrated example ofFIG. 6, the example impressions corrector238 ofFIG. 2 applies the IA factors502 to the DT-corrected impression counts608 to determine the misattribution-corrected impression counts604. That is, the example impressions corrector238 increases a DT-correctedimpression count608 for a demographic group based on a correspondingIA factor502 for the same demographic group to calculate a corresponding misattribution-corrected impressions count604. In particular, the example impressions corrector238 multiples anIA factor502 for a demographic group by the total DP DT-corrected impressions count612 summed across all of the demographic groups to determine a number of adjustment impressions by which to adjust the DP DT-corrected impressions count608 for the same demographic group corresponding to the selected IA factor502 (e.g., (adjustment impressions)=(IA factor)×(total cross-demographic DP DT-corrected impressions)). The example impressions corrector238 then adds the calculated adjustment impressions to the corresponding DP DT-corrected impressions count608 for the same demographic group to determine a corresponding misattribution-corrected impressions count604 (e.g., (misattribution-corrected impressions count)=(DP DT-corrected impressions count608)+(adjustment impressions)).
• For example, to determine the misattribution-correctedimpressions604 corresponding to the F<50 demographic group, the example impressions corrector238 ofFIG. 2 multiples theIA factor502 of 11.11% for the F<50 demographic group by the total DP DT-corrected impressions count612 of 710,000 to calculate the adjustment impressions of 78,888. The example impressions corrector238 then adds the 78,888 adjustment impressions to the corresponding DP DT-corrected impressions count608 of 210,000 for the F<50 demographic group to calculate the misattribution-corrected impressions count604 of 288,889 for the F<50 demographic group.
• To determine the misattribution-correctedimpressions604 for the M<50 demographic group, the example impressions corrector238 ofFIG. 2 multiples theIA factor502 of −11.11% for the M<50 demographic group by the total DP DT-corrected impressions count612 of 710,000 to calculate the adjustment impressions of −78,888. The example impressions corrector238 then adds the −78,888 adjustment impressions to (or subtracts 78,888 from) the corresponding DP DT-corrected impressions count608 of 165,000 for the M<50 demographic group to calculate the misattribution-corrected impressions count604 of 86,111 for the M<50 demographic group.
• An alternative technique to determine the misattribution-corrected unique audience sizes involves using impressions frequency values as described in connection withFIG. 7.FIG. 7 illustrates an example table700 with misattribution-corrected unique audience values702 and misattribution-corrected impression counts604 determined based on the IA factors502 ofFIG. 5 and impression frequencies706 for different demographic groups. As used herein, impressions frequency is a number of total impressions (e.g., a DP DT-correctedimpression count608 ofFIG. 6) divided by a quantity of unique audience members (e.g., a DT-correctedUA size value606 ofFIG. 6) (e.g., frequency=impressions count/UA). For example, for the F<50 demographic group, the database proprietor impressions frequency706 is 3.33, which is calculated by dividing 210,000 DP DT-corrected impressions by 63,000 DP DT-corrected UA. In the illustrated example ofFIG. 7, after the example impressions corrector238 determines the misattributions-correctedimpressions604 based on the IA factors502 as described above in connection withFIG. 6, the exampleunique audience corrector236 divides the misattribution-correctedimpressions604 of 288,889 for the F<50 demographic group by the DP frequency of 3.33 (for the F<50 demographic group) to calculate a misattribution-correctedUA size702 of 86,667. The frequency-based approach to determining misattribution-corrected impressions704 preserves the impressions frequencies for the demographic groups.
• As shown inFIGS. 6 and 7, the misattribution-correctedUA sizes602 ofFIG. 6 are different from the misattribution-correctedUA sizes702 ofFIG. 7. In determining whether to use the AA factor approach described above in connection withFIG. 6 or the impressions frequency approach described in connection withFIG. 7 to determine misattribution-corrected UA sizes, both approaches can be applied over multiple iterations on test data for which true UA sizes are known. The approach that produces the most accurate misattribution-corrected UA sizes relative to the true UA sizes can then be selected for use on real impression data. Alternatively, the impression frequency approach may be selected if a party wishes to preserve impression frequency even if the accuracies of resulting misattribution-corrected UA sizes are not optimal.
• An example advantage of example misattribution adjustment techniques disclosed herein is that the total DP DT-corrected impressions count612 (e.g., 710,000 impressions inFIGS. 6 and 7) remains the same after correcting the data for misattribution errors. That is, impressions are not changed, but are instead redistributed. For example, as shown inFIGS. 6 and 7, a total misattribution-corrected impressions count614 across all demographic groups is 710,000, which is equal to the DP DT-corrected impressions count612 of 710,000.
• FIG. 8 is a flow diagram representative of machine readable instructions that may be executed to implement themisattribution corrector202 ofFIG. 2 to determine the AA factors402 ofFIGS. 4 and 6, the IA factors502 ofFIGS. 5,6, and7, the misattribution-correctedunique audience sizes602 ofFIG. 6, the misattribution-correctedunique audience sizes702 ofFIG. 7, and the misattribution-corrected impression counts604 ofFIGS. 6 and 7. In this example, the machine readable instructions comprise one or more programs for execution by a processor such as theprocessor912 shown in theexample processor platform900 discussed below in connection withFIG. 9. The program(s) may be embodied in software stored on a tangible computer readable storage medium such as a CD-ROM, a floppy disk, a hard drive, a digital versatile disk (DVD), a Blu-ray disk, or a memory associated with theprocessor912, but the entire program and/or parts thereof could alternatively be executed by a device other than theprocessor912 and/or embodied in firmware or dedicated hardware. Further, although the example program(s) is/are described with reference to the flowchart illustrated inFIG. 8, many other methods of implementing theexample misattribution corrector202 may alternatively be used. For example, the order of execution of the blocks may be changed, and/or some of the blocks described may be changed, eliminated, or combined.
• As mentioned above, the example process(es) ofFIG. 8 may be implemented using coded instructions (e.g., computer and/or machine readable instructions) stored on a tangible computer readable storage medium such as a hard disk drive, a flash memory, a read-only memory (ROM), a compact disk (CD), a digital versatile disk (DVD), a cache, a random-access memory (RAM) and/or any other storage device or storage disk in which information is stored for any duration (e.g., for extended time periods, permanently, for brief instances, for temporarily buffering, and/or for caching of the information). As used herein, the term tangible computer readable storage medium is expressly defined to include any type of computer readable storage device and/or storage disk and to exclude propagating signals and to exclude transmission media. As used herein, “tangible computer readable storage medium” and “tangible machine readable storage medium” are used interchangeably. Additionally or alternatively, the example process(es) ofFIG. 8 may be implemented using coded instructions (e.g., computer and/or machine readable instructions) stored on a non-transitory computer and/or machine readable medium such as a hard disk drive, a flash memory, a read-only memory, a compact disk, a digital versatile disk, a cache, a random-access memory and/or any other storage device or storage disk in which information is stored for any duration (e.g., for extended time periods, permanently, for brief instances, for temporarily buffering, and/or for caching of the information). As used herein, the term non-transitory computer readable medium is expressly defined to include any type of computer readable storage device and/or storage disk and to exclude propagating signals and to exclude transmission media. As used herein, when the phrase “at least” is used as the transition term in a preamble of a claim, it is open-ended in the same manner as the term “comprising” is open ended.
• The example flow diagram ofFIG. 8 is shown as two phases including an example adjustmentfactors development phase802 and an examplemisattribution correction phase804. During the adjustment factorsdevelopment phase802, the misattribution corrector202 (FIG. 2) determines the AA factors402 (FIGS. 4 and 6) and the IA factors502 (FIGS. 5 and 6) for different demographic groups based on development or test impressions such as the impressions shown in table300 ofFIG. 3. During themisattribution correction phase804, themisattribution corrector202 corrects aggregate impression data (e.g., unique audience measures and total impression counts) generated based on impressions collected by the database proprietor116 (and/or one or more other database proprietors). For example, themisattribution corrector202 uses the AA factors402 and/or the IA factors502 to determine the misattribution corrected UA size values602 ofFIG. 6, the misattribution-corrected UA size values702 ofFIG. 7, and/or the misattribution-corrected impression counts604 ofFIGS. 6 and 7 for different demographic groups. In some examples, themisattribution correction phase804 may begin immediately after the adjustment factorsdevelopment phase802. In other examples, themisattribution correction phase804 may begin after a significant amount of time (e.g., hours, days, weeks, etc.) has passed following the completion of the adjustment factorsdevelopment phase802. In some examples, the adjustment factorsdevelopment phase802 and themisattribution correction phase804 may be implemented as part of a same program. In other examples, the adjustment factorsdevelopment phase802 and themisattribution correction phase804 may be implemented as two separate programs.
• The example adjustmentfactors development phase802 ofFIG. 8 begins atblock806 at which theAME impressions collector218 collects impressions from theclient device102. For example, theAME impressions collector218 collects impressions using the techniques described above in connection withFIG. 2. TheDP impressions collector230 obtains development impression records from thedatabase proprietor116 that correspond to AME panelists that are also subscribers of the database proprietor116 (block808). Themisattribution corrector202 selects a demographic group (block810). For example, themisattribution corrector202 selects one of the demographic groups ofFIGS. 4-7. The example impressions adjustment factor determiner234 (FIG. 2) determines anIA factor502 for the selected demographic group (block812). For example, the impressionsadjustment factor determiner234 determines theIA factor502 usingEquation 2 above and/or the technique described above in connection withFIG. 5.
• The example unique audience adjustment factor determiner232 (FIG. 2) determines anAA factor402 for the selected demographic group (block814). For example, the unique audienceadjustment factor determiner232 determines theAA factor402 usingEquation 1 above and/or the technique described above in connection withFIG. 4. Themisattribution corrector202 determines whether there is another demographic group for which to determine adjustment factors (block816). If there is another demographic group, control returns to block810. If there is not another demographic group, the adjustment factorsdevelopment phase802 ends. In the illustrated example, after the adjustment factorsdevelopment phase802 ends, themisattribution correction phase804 begins based on the IA factors502 and the AA factors402 determined during the adjustment factorsdevelopment phase802. In some examples, the adjustment factorsdevelopment phase802 is repeated from time to time (e.g., after a number of days, weeks, months, etc.) to update the IA factors502 and/or the AA factors402. For example, the ability of thedatabase proprietor116 to identify subscribers may change (e.g., increased or decreased accuracy) from time to time. As such, to increase the likelihood that the IA factors502 and the AA factors402 reflect such changes, the adjustment factorsdevelopment phase802 can be repeated from time to time.
• In themisattribution correction phase804, theDP impressions collector230 obtains the DP DT-corrected unique audience sizes606 (FIG. 6) and DP DT-corrected impression counts608 (FIG. 6) from the database proprietor116 (block818). Themisattribution corrector202 selects a demographic group (block820). For example, themisattribution corrector202 selects one of the demographic groups ofFIGS. 4-7. The example impressions corrector238 (FIG. 2) determines a misattribution-corrected impressions count604 (FIGS. 6 and 7) based on theIA factor502 for the selected demographic group (block822). For example, the impressions corrector238 can determine the misattribution-corrected impressions count604 as described above in connection withFIG. 6.
• Themisattribution corrector202 determines whether to use impressions frequency to determine a misattribution-corrected unique audience size (block824). For example, themisattribution corrector202 may check a configuration setting in a file, a program, and/or a hardware setting indicating whether to determine a misattribution-corrected unique audience size based on an impressions frequency706 (FIG. 7). If themisattribution corrector202 determines that it should determine a misattribution-corrected unique audience size based on an impressions frequency706, themisattribution corrector202 determines an impressions frequency706 for the selected demographic profile (block826). For example, themisattribution corrector202 may determine the impressions frequency706 for the selected demographic profile as described above in connection withFIG. 7. If themisattribution corrector202 determines that it should not use an impressions frequency706 to determine a misattribution-corrected unique audience size, control advances to block828 without determining an impressions frequency706. In some examples, the impressions frequency706 is determined by thedatabase proprietor116 and provided by thedatabase proprietor116 to themisattribution corrector202 via theDP impressions collector230. In such examples, themisattribution corrector202 does not need to determine the impressions frequency706.
• Atblock828, the example unique audience corrector236 (FIG. 2) determines the misattribution-corrected UA size for the selected demographic group (block828). For example, if themisattribution corrector202 determined atblock824 that the impressions frequency706 is to be used to determine the misattribution-corrected UA size702 (FIG. 7) for the selected demographic group, theunique audience corrector236 determines the misattribution-correctedUA size702 based on the impressions frequency706 ofblock826 as described above in connection withFIG. 7. Alternatively atblock828, if themisattribution corrector202 determined atblock824 that the impressions frequency706 is not to be used to determine the misattribution-corrected UA size602 (FIG. 6) for the selected demographic group, theunique audience corrector236 determines the misattribution-correctedUA size602 based on the AA factor402 (FIG. 4) of the selected demographic group as described above in connection withFIG. 6.
• Themisattribution corrector202 then determines whether there is another demographic group for which misattribution-adjusted impression counts or misattribution-adjusted UA sizes are to be determined (block830). If there is another demographic group, control returns to block820. Otherwise, the example program ofFIG. 8 ends.
• FIG. 9 is a block diagram of anexample processor platform900 capable of executing the instructions ofFIG. 9 to implement themisattribution corrector202 ofFIG. 2. Theprocessor platform900 can be, for example, a server, a personal computer, or any other type of computing device.
• Theprocessor platform900 of the illustrated example includes aprocessor912. Theprocessor912 of the illustrated example is hardware. For example, theprocessor912 can be implemented by one or more integrated circuits, logic circuits, microprocessors or controllers from any desired family or manufacturer.
• In the illustrated example, theprocessor912 implements theexample misattribution corrector202, the exampleAME impressions collector218, the exampleDP impressions collector230, the example audienceadjustment factor determiner232, the example impressionsadjustment factor determiner234, the exampleunique audience corrector236, and/or the example impressions corrector238 described above in connection withFIG. 2.
• Theprocessor912 of the illustrated example includes a local memory913 (e.g., a cache). Theprocessor912 of the illustrated example is in communication with a main memory including avolatile memory914 and anon-volatile memory916 via abus918. Thevolatile memory914 may be implemented by Synchronous Dynamic Random Access Memory (SDRAM), Dynamic Random Access Memory (DRAM), RAMBUS Dynamic Random Access Memory (RDRAM) and/or any other type of random access memory device. Thenon-volatile memory816 may be implemented by flash memory and/or any other desired type of memory device. Access to themain memory914,916 is controlled by a memory controller.
• In the illustrated example, thelocal memory913 stores the example modifiedID map228 described above in connection withFIG. 2. In other examples any one or more of thelocal memory913, therandom access memory914, the read onlymemory916, and/or amass storage device928 may store the example modifiedID map228.
• Theprocessor platform900 of the illustrated example also includes aninterface circuit920. Theinterface circuit920 may be implemented by any type of interface standard, such as an Ethernet interface, a universal serial bus (USB), and/or a PCI express interface.
• In the illustrated example, one ormore input devices922 are connected to theinterface circuit920. The input device(s)922 permit(s) a user to enter data and commands into theprocessor912. The input device(s) can be implemented by, for example, an audio sensor, a microphone, a camera (still or video), a keyboard, a button, a mouse, a touchscreen, a track-pad, a trackball, isopoint and/or a voice recognition system.
• One ormore output devices924 are also connected to theinterface circuit920 of the illustrated example. Theoutput devices924 can be implemented, for example, by display devices (e.g., a light emitting diode (LED), an organic light emitting diode (OLED), a liquid crystal display, a cathode ray tube display (CRT), a touchscreen, a tactile output device, a light emitting diode (LED), a printer and/or speakers). Theinterface circuit920 of the illustrated example, thus, typically includes a graphics driver card, a graphics driver chip or a graphics driver processor.
• Theinterface circuit920 of the illustrated example also includes a communication device such as a transmitter, a receiver, a transceiver, a modem and/or network interface card to facilitate exchange of data with external machines (e.g., computing devices of any kind) via a network926 (e.g., an Ethernet connection, a digital subscriber line (DSL), a telephone line, coaxial cable, a cellular telephone system, etc.).
• Theprocessor platform900 of the illustrated example also includes one or moremass storage devices928 for storing software and/or data. Examples of suchmass storage devices928 include floppy disk drives, hard drive disks, compact disk drives, Blu-ray disk drives, RAID systems, and digital versatile disk (DVD) drives.
• Coded instructions932 include the machine readable instructions ofFIG. 8 and may be stored in themass storage device928, in thevolatile memory914, in thenon-volatile memory916, and/or on a removable tangible computer readable storage medium such as a CD or DVD.
• From the foregoing, it will be appreciate that methods, apparatus and articles of manufacture have been disclosed which enhance the operations of a computer to improve the accuracy of impression-based data such as unique audience and impression counts so that computers and processing systems therein can be relied upon to produce audience analysis information with higher accuracies. In some examples, computer operations can be made more efficient based on the above equations and techniques for determining IA factors, AA factors, misattribution-corrected unique audience sizes, and misattribution-corrected impression counts. That is, through the use of these processes, computers can operate more efficiently by relatively quickly determining parameters and applying those parameters through the above disclosed techniques to determine the misattribution-corrected data. For example, using example processes disclosed herein, a computer can more efficiently and effectively identify misattribution errors (e.g., themisattribution error308 of FIG.3) in development or test data logged by theAME114 and thedatabase proprietor116 without using large amounts of network communication bandwidth (e.g., conserving network communication bandwidth) and without using large amounts of computer processing resources (e.g., conserving processing resources) to communicate with individual online users to request survey responses about their online media access habits and without needing to rely on such survey responses from such online users. Survey responses from online users can be inaccurate due to inabilities or unwillingness of users to recollect online media accesses. Survey responses can also be incomplete, which could require additional processor resources to identify and supplement incomplete survey responses. As such, examples disclosed herein more efficiently and effectively determine misattribution-corrected data. Such misattribution-corrected data is useful in subsequent processing for identifying exposure performances of different media so that media providers, advertisers, product manufacturers, and/or service providers can make more informed decisions on how to spend advertising dollars and/or media production and distribution dollars.
• Furthermore, example methods, apparatus, and/or articles of manufacture disclosed herein identify and overcome inaccuracies in impressions and/or aggregate impression-based data provided by database proprietors. For example, example methods, apparatus, and/or articles of manufacture disclosed herein overcome the technical problem of counting impressions and determining unique audiences of media on media devices that are shared by multiple people. Example methods, apparatus, and/or articles of manufacture disclosed herein solve this problem without forcing such media devices to be used by only a single person and without forcing people to always login to their subscriber accounts of database proprietors. By not forcing logins into database proprietor accounts, examples disclosed herein do not force additional network communications to be employed, thus, reducing network traffic.
• Although certain example methods, apparatus and articles of manufacture have been disclosed herein, the scope of coverage of this patent is not limited thereto. On the contrary, this patent covers all methods, apparatus and articles of manufacture fairly falling within the scope of the claims of this patent.

Claims (32)

The status of the claims:

1. A method, comprising:

receiving, at a first internet domain, a first request from a client device, the first request indicative of access to media at the client device;

sending, from the first internet domain, a response to the client device, the response to instruct the client device to send a second request to a second internet domain, the second request to be indicative of the access to the media at the client device;

determining an impressions adjustment factor for a first demographic group based on first impressions reported by the client device to the first internet domain and second impressions reported by the client device to the second internet domain, the first and second impressions corresponding to the same media accessed on the client device; and

determining a misattribution-corrected impressions count for the first demographic group based on the impressions adjustment factor and based on a second impressions count determined at the second internet domain for the first demographic group, the second impressions count having an error based on some of the second impressions being misattributed at the second internet domain to the first demographic group when the some of the second impressions correspond to a second demographic group.

2. A method as defined inclaim 1, wherein determining the misattribution-corrected impression count comprises shifting an impression from the second impressions count corresponding to the first demographic group to a third impressions count corresponding to the second demographic group based on the impressions adjustment factor.

3. A method as defined inclaim 1, wherein the first impressions are reported by the client device to an audience measurement entity at the first internet domain that does not provide the media to the client device, and a user of the client device is a panel member of the audience measurement entity.

4. A method as defined inclaim 1, wherein the second impressions are reported by the client device to a social network service at the second internet domain to which a user of the client device is subscribed.

5. A method as defined inclaim 1, wherein determining the impressions adjustment factor comprises:

subtracting a first unique audience size determined by an audience measurement entity at the first internet domain based on the first impressions from a second unique audience size determined by a database proprietor at the second internet domain based on the second impressions to generate a difference; and

dividing the difference by a total impressions count of the first impressions.

6. A method as defined inclaim 1, wherein the impressions adjustment factor is to correct impression quantities having inaccuracies due to impressions incorrectly attributed to demographic data not corresponding to persons corresponding to the impressions.

7. A method as defined inclaim 1, wherein the error in the second impressions count is based on an entity at the second internet domain incorrectly identifying a user of the client device as belonging to the first demographic group when the user belongs to the second demographic group, the misattribution-corrected impressions count comprising fewer impressions than the second impression count based on shifting an impression corresponding to the user from the second impressions count corresponding to the first demographic group to a third impressions count corresponding to the second demographic group based on the impressions adjustment factor.

8. A method as defined inclaim 1, wherein the misattribution-corrected impressions count is determined based on the impressions adjustment factor without communicating with individual online users about their online media access activities and without using survey responses from the online users to determine the error.

9. A method as defined inclaim 8, further comprising conserving network communication bandwidth by not communicating with individual online users about their online media access activities and by not requesting survey responses from the online users to determine the error.

10. A method as defined inclaim 8, further comprising conserving computer processing resources by not communicating with individual online users about their online media access activities and by not requesting survey responses from the online users to determine the error.

11. An apparatus, comprising:

an impression collector to:

receive, at a first internet domain, a first request from a client device, the first request indicative of access to media at the client device; and

send, from the first internet domain, a response to the client device, the response to instruct the client device to send a second request to a second internet domain, the second request to be indicative of the access to the media at the client device;

an impressions adjustment factor determiner to determine an impressions adjustment factor for a first demographic group based on first impressions reported by the client device to the first internet domain and second impressions reported by the client device to the second internet domain, the first and second impressions corresponding to the same media accessed on the client device; and

an impressions corrector to determine, via a processor, a misattribution-corrected impressions count for the first demographic group based on the impressions adjustment factor and based on a second impressions count determined at the second internet domain for the first demographic group, the second impressions count having an error based on some of the second impressions being misattributed at the second internet domain to the first demographic group when the some of the second impressions correspond to a second demographic group.

12. An apparatus as defined inclaim 11, wherein the impressions corrector is to determine the misattribution-corrected impressions count by shifting an impression from the second impressions count corresponding to the first demographic group to a third impressions count corresponding to the second demographic group based on the impressions adjustment factor.

13. An apparatus as defined inclaim 11, wherein the first impressions are reported by the client device to an audience measurement entity at the first internet domain that does not provide the media to the client device, and a user of the client device is a panel member of the audience measurement entity.

14. An apparatus as defined inclaim 11, wherein the second impressions are reported by the client device to a social network service at the second internet domain to which a user of the client device is subscribed.

15. An apparatus as defined inclaim 11, wherein the impressions adjustment factor determiner is to:

determine the impressions adjustment factor by subtracting a first unique audience size determined by an audience measurement entity at the first internet domain based on the first impressions from a second unique audience size determined by a database proprietor at the second internet domain based on the second impressions to generate a difference; and

divide the difference by a total impressions count of the first impressions.

16. An apparatus as defined inclaim 11, wherein the impressions adjustment factor is to correct impression quantities having inaccuracies due to impressions incorrectly attributed to demographic data not corresponding to persons corresponding to the impressions.

17. An apparatus as defined inclaim 11, wherein the error in the second impressions count is based on an entity at the second internet domain incorrectly identifying a user of the client device as belonging to the first demographic group when the user belongs to the second demographic group, the misattribution-corrected impressions count comprising fewer impressions than the second impression count based on shifting an impression corresponding to the user from the second impressions count corresponding to the first demographic group to a third impressions count corresponding to the second demographic group based on the impressions adjustment factor.

18. An apparatus as defined inclaim 11, wherein the impressions corrector determines the misattribution-corrected impressions based on the impressions adjustment factor without communicating with individual online users about their online media access activities and without using survey responses from the online users to determine the error.

19. An apparatus as defined inclaim 18, wherein the impressions corrector determining the misattribution-corrected impressions conserves network communication bandwidth by not communicating with individual online users about their online media access activities and by not requesting survey responses from the online users to determine the error.

20. An apparatus as defined inclaim 18, wherein the impressions corrector determining the misattribution-corrected impressions conserves computer processing resources by not communicating with individual online users about their online media access activities and by not requesting survey responses from the online users to determine the error.

21-30. (canceled)

31. A method, comprising:

receiving, at a first internet domain, a first request from a client device, the first request indicative of access to media at the client device;

sending, from the first internet domain, a response to the client device, the response to instruct the client device to send a second request to a second internet domain, the second request to be indicative of the access to the media at the client device;

determining an audience adjustment factor for a demographic group based on first impressions reported by the client device to the first internet domain and second impressions reported by the client device to the second internet domain, the first and second impressions corresponding to the same media accessed on the client device; and

determining a misattribution-corrected unique audience size for the demographic group based on the audience adjustment factor and based on a second unique audience size determined at the second internet domain for the demographic group, the second unique audience size having an error based on third impressions misattributed at the second internet domain to the demographic group when the third impressions correspond to another demographic group.

32. A method as defined inclaim 31, wherein determining the audience adjustment factor comprises dividing a third unique audience size corresponding to the first impressions by a fourth unique audience size corresponding to the second impressions.

33. A method as defined inclaim 31, wherein determining the misattribution-corrected unique audience size for the demographic group comprises dividing the second unique audience size by the audience adjustment factor.

34. A method as defined inclaim 31, wherein the first impressions are reported by the client device to an audience measurement entity at the first internet domain that does not provide the media to the client device, and a user of the client device is a panel member of the audience measurement entity.

35. A method as defined inclaim 31, wherein the second impressions are reported by the client device to a social network service at the second internet domain to which a user of the client device is subscribed.

36. A method as defined inclaim 31, wherein the audience adjustment factor is to correct unique audience size values having inaccuracies due to impressions incorrectly attributed to demographic data not corresponding to persons corresponding to the impressions.

37. A method as defined inclaim 31, wherein the error in the second unique audience size is based on an entity at the second internet domain incorrectly identifying a user of the client device as belonging to the demographic group when the user belongs to the another demographic group, the misattribution-corrected unique audience size being different than the second unique audience size based on dividing the second unique audience size by the audience adjustment factor.

38. A method as defined inclaim 31, wherein the misattribution-corrected unique audience size is determined based on the audience adjustment factor without communicating with individual online users about their online media access habits and without using survey responses from the online users to determine the error.

39. A method as defined inclaim 38, further comprising conserving network communication bandwidth by not communicating with individual online users about their online media access habits and by not requesting survey responses from the online users to determine the error.

40. A method as defined inclaim 38, further comprising conserving computer processing resources by not communicating with individual online users about their online media access habits and by not requesting survey responses from the online users to determine the error.

41-60. (canceled)

Images